The main goal of the project was to better understand the impact to a nuclear power plant due to the unavailability of critical infrastructure. We evaluated the use of rare event analysis to establish rare event occurrences in the vicinity of the plant. For the purpose of this report, rare events were considered extreme scenarios of natural disasters. The initial data was acquired by viewing STP FSARs and topographical maps of the region. The two specific factors that were extracted by analyzing the initial data of the plant design and location were the road networks and the electrical grid system. These two variables were then analyzed in light of the rare event analysis model that was used as a building block for this project. The rare event analysis also provided information regarding the types of data and distributions that were likely to be seen as a result. For the road networks, the road layout around the plant was mapped and the relevant data was used to consider the possible routes to the plants in case of a rare event. Data regarding LOOP Events were gathered mainly from NRC Documents and Institutional Sources. These data were used to analyze how many events occurred per year and the downtime associated with the rare event.
Adaptive Multilevel Splitting for Rare Event Analysis